| To overcome the difficulty in managing oilfields due to their dispersion,the complex operating environment of downhole electric submersible pump,the high cost of working condition diagnostic instrument and the incomprehensive judgment of current card method.A real-time diagnostic method for the working condition of the electric submersible pump is proposed,which is based on wellhead electrical parameters and production parameters and combined with the characteristics of new developed tubular flowmeter for real-time measurement of gas-liquid production and oil pressure.The as-proposed method can reflect downhole production status,equipment failure and reservoir condition through on-well parameters.The main work and research results are as follows:Based on the historical data of wellhead electrical parameters and production parameters of multi-wells in block and other rated process parameters of oil wells,different characteristics of various working conditions of electric submersible pump are analyzed,and a method of working condition diagnosis of electric submersible pump based on wellhead electrical parameters and production parameters is proposed,and the software and hardware design of working condition diagnosis system is completed.The multi-source signal acquisition of electric submersible pump has been accomplished.RTU(Remote Terminal Unit)is used to realize the real-time acquisition of wellhead electrical parameters and production parameters and the wireless remote transmission of 4G.The data of wellhead technical parameters and history data are collected and recorded,which provides a data basis for the diagnosis of ESP operating conditions.Combining with the rated process parameters and historical data of oil wells,the method of working condition diagnosis of submersible electric pump based on wellhead electrical parameters and production parameters is put forward by combining the method of feature matching and prior knowledge logic,and the working condition diagnosis algorithm and software design are completed.Simple time-domain features such as mean,variance,deviation and extremum are used to realize real-time diagnosis of seven kinds of simple working conditions,such as gas impact,gas locking,underload and pipe string leakage.In view of complex working conditions,the historical working conditions of different wells and wells in the same area are synthesized.As the initial standard operating conditions of the well,the correlation detection matching between the standard operating conditions and the measured signals is carried out,and the real-time diagnosis of complex working conditions such as gas lock and insufficient liquid supply is realized.With the emergence of real-time complex working conditions in other wells and in this well,the recursive least squares method is used to realize the self-adaptive correction of the standard working condition characteristics and update the working condition feature library in real time.Through the prediction of complex working conditions,the further development of fault working condition is effectively prevented.According to dynamic grey model and correlation analysis method,the current data of complex working conditions are predicted,and fault information is obtained in advance,which can effectively prevents the further development of fault working conditions.The multi-scale transformation is used to reduce the difference caused by the real-time working condition characteristics in the magnitude and time scales,thus improving the recognition rate of working condition diagnosis.The working condition diagnosis system is tested in laboratory and field.The results show that the system runs smoothly,and achieves data acquisition and 4G remote transmission.Utillizing the existing and simulated working condition data for off-line testing,the test results indicate that the diagnostic rate of the working condition is as high as 98%,which demonstrates that the forecasting model can effectively predict the fault condition. |
PDF Full Text Request